domains / families / research / andrej-karpathy

andrej-karpathy

authentic

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

When to use

• Use when writing, reviewing, or refactoring code with an LLM.
• Use when a change needs to stay surgical and avoid speculative abstractions.
• Use when assumptions, tradeoffs, and verification criteria should be made explicit.
• Use when code has become overcomplicated and needs to be simplified.

When not to use

• These guidelines are behavioral guardrails, not a replacement for project-specific architecture or style rules.
• For emergency fixes, prioritize the smallest verified correction over extensive planning.
• For exploratory prototypes, some caution can be relaxed, but assumptions and verification should still be explicit.

Skill page

Source